Explore the directory

An edge-native mixture-of-experts serving engine that runs frontier-scale open-weight language models on consumer GPUs, with CPU-GPU co-execution, semantic caching and Anthropic/OpenAI-compatible APIs.

• Inference

A Python library for getting reliable, typed structured output from any large language model, built on Pydantic for validation and automatic retries instead of hand-written JSON parsing.

• Libraries

Open-source infrastructure for running AI-generated code in secure, isolated cloud sandboxes, with JavaScript and Python SDKs, a code-interpreter mode and a desktop SDK for computer-use agents.

• Containers

A Chinese-language, hands-on tutorial that teaches beginner developers to build large language model applications, from calling LLM APIs through building a knowledge-base RAG assistant to iterative evaluation.

• Tutorials

A framework for few-shot evaluation of language models across a large, config-driven library of benchmark tasks, with pluggable model backends, filters and metrics.

• AI evals

A minimal, self-evolving autonomous agent framework built from about 3,000 lines of code and nine atomic tools, giving a language model system-level control of a local computer and growing new skills from every task.

• Agent frameworks

A framework that automates high-value research-and-development work in AI, letting AI agents drive the model and data side of AI research, from ideation through implementation to evaluation.

• Agent frameworks
• Models

A community-maintained native desktop app for installing, configuring and chatting with Hermes Agent, a self-improving AI assistant with tool use and multi-platform messaging.

• Agent frameworks

A Python framework for building realtime, multimodal voice AI agents that run on servers, mixing speech-to-text, language models, text-to-speech and telephony into conversational participants.

• Agent frameworks
• Speech

A lightweight, agent-driven memory system that gives coding assistants and other AI agents one shared memory wiki across sessions, agents and devices, and distills reusable skills from their history.

• Agent memory

A self-hosted identity and access management server with a web console, speaking OAuth 2.0, OIDC, SAML, CAS, LDAP, SCIM, WebAuthn, TOTP and MCP, and connecting to providers such as Google Workspace and GitHub.

• Auth

An autonomous multi-agent research pipeline that takes a research idea through experiment design, execution and paper writing, with an optional human-in-the-loop co-pilot mode.

• Agent frameworks

A Model Context Protocol server that bridges AI assistants such as Claude, Codex and Cursor with the Unity Editor, giving them tools to manage assets, control scenes, edit scripts, run tests and automate game-development workflows.

• MCP servers

An engineering framework that persists specs, tasks and project memory into a repository so any AI coding agent works to the team's own conventions across sessions, instead of starting from scratch each time.

• Coding agents
• Agent memory

A self-hosted, open-source WhatsApp API gateway with a pluggable architecture for its database, storage, and cache backends, and session- and chat-scoped API keys.

• APIs

A desktop agentic IDE for running Claude Code, Codex or other CLI coding agents in parallel, with per-task Git worktrees, a built-in diff viewer, in-app browser previews and live monitoring of every agent.

• Coding agents

An SDK that provides a Python API for defining large language models and running them efficiently on NVIDIA GPUs, with specialized kernels, an optimized runtime and components for building custom inference servers.

• Inference

An RNN architecture that reaches transformer-level language-model performance while training in parallel like a transformer, with linear time, constant memory, no key-value cache and unlimited context length.

• Models

A cross-platform desktop workspace for Claude Code and other coding agents, with multi-session management, Git worktrees, diff review, a skill marketplace and remote access through chat apps.

• Coding agents

A framework for building local, private, enterprise retrieval-augmented generation applications, pairing a catalog of 300+ small, optimized language models with document parsing and knowledge-base pipelines.

• RAG
• Models

An independent, open-source research reimplementation of the Claude Mythos model architecture, a recurrent-depth transformer with a looped block and a sparse mixture-of-experts feed-forward network.

• Models

An open-source terminal, SSH, SFTP, FTP, Telnet, serial-port, RDP, VNC and Spice client that runs across desktop, mobile and even older or less common operating systems and CPU architectures.

• Terminals

A platform and toolkit for building and deploying chatbots and assistants backed by large language models, including a CLI and SDK for creating custom integrations and bots.

• Backend
• Workflow automation

An application for building datasets to fine-tune, retrieval-augment and evaluate large language models, with document parsing, text splitting, question generation and data cleaning built in.

• Models
• Data engineering